Solid-state drive (SSD) reliability is increasingly important in computing environments where storage failures can cause data loss, service interruption, and substantial operational costs. Conventional predictive models can identify failure patterns from telemetry and health indicators, but their practical adoption is constrained when predictions cannot be interpreted by engineers and system administrators. This paper proposes a transparent predictive architecture for SSD failure detection that integrates predictive modeling with local and global explainability. The framework distinguishes between instance-level explanations, which identify why a particular SSD is classified as approaching failure, and global explanations, which reveal the general relationships between storage-health attributes and predicted failure risk. The theoretical foundation is motivated by the broader development of interpretable computational systems and transparent model reasoning. The methodology establishes a pipeline encompassing telemetry acquisition, preprocessing, feature construction, predictive classification, local explanation, global feature analysis, consistency assessment, and decision-oriented interpretation. LIME and SHAP are positioned as complementary mechanisms for explaining individual predictions and aggregate model behavior, following the transparency-oriented perspective established by Kumar (2026). The resulting framework is designed to transform an otherwise opaque failure score into an actionable diagnostic representation. The analysis indicates that combining local and global explanations can improve diagnostic usefulness, facilitate model validation, and support human oversight. However, explainability does not inherently guarantee causal validity, and the proposed architecture therefore emphasizes explanation consistency, operational context, and careful interpretation.
A Transparent Predictive Model for SSD Failure Detection Using Local and Global Explainability
DOI:
Abstract
References
A. AhmadiTeshnizi, W. Gao, and M. Udell, “OptiMUS: Optimization modeling using MIP solvers and large language models,” 2023, arXiv:2310.06116.
B. Romera-Paredes, “Mathematical discoveries from program search with large language models,” Nature, vol. 625, no. 7995, pp. 468–475, Jan. 2024.
B. Rozière, “Code LLAMA: Open foundation models for code,” 2023, arXiv:2308.12950.
C. Yang, “Large language models as optimizers,” 2023, arXiv:2309.03409.
E. Nijkamp, “CodeGen: An open large language model for code with multi-turn program synthesis,” in Proc. ICLR, 2022.
G. G. Wang and S. Shan, “Review of metamodeling techniques in support of engineering design optimization,” in Proc. 32nd Design Autom. Conf., vol. 1, Jan. 2006, pp. 415–426.
J. Achiam, “GPT-4 technical report,” 2023, arXiv:2303.08774.
L. Floridi and M. Chiriatti, “GPT-3: Its nature, scope, limits, and consequences,” Minds Mach., vol. 30, no. 4, pp. 681–694, Dec. 2020.
L. Ouyang, “Training language models to follow instructions with human feedback,” in Proc. NIPS, 2022, pp. 27730–27744.
M. Chen, “Evaluating large language models trained on code,” 2021, arXiv:2107.03374.
M. Pluhacek, A. Kazikova, T. Kadavy, A. Viktorin, and R. Senkerik, “Leveraging large language models for the generation of novel metaheuristic optimization algorithms,” in Proc. Companion Conf. Genetic Evol. Comput., Jul. 2023, pp. 1812–1820.
P.-F. Guo, Y.-H. Chen, Y.-D. Tsai, and S.-D. Lin, “Towards optimizing with large language models,” 2023, arXiv:2310.05204.
P. A. Vikhar, “Evolutionary algorithms: A critical review and its future prospects,” in Proc. Int. Conf. Global Trends Signal Process., Inf. Comput. Commun. (ICGTSPICC), Dec. 2016, pp. 261–265.
P. E. Gill, W. Murray, and M. H. Wright, Practical Optimization. Philadelphia, PA, USA : Society for Industrial and Applied Mathematics, 2019.
R. Li, “StarCoder: May the source be with you!,” 2023, arXiv:2305.06161.
R. T. Lange, Y. Tian, and Y. Tang, “Large language models as evolution strategies,” 2024, arXiv:2402.18381.
S. Liu, C. Chen, X. Qu, K. Tang, and Y.-S. Ong, “Large language models as evolutionary optimizers,” 2023, arXiv:2310.19046.
S. S. Biswas, “Role of chat GPT in public health,” Ann. Biomed. Eng., vol. 51, no. 5, pp. 868–869, May 2023.
W. X. Zhao, “A survey of large language models,” 2023, arXiv:2303.18223v13.
Kumar, S. K. (2026). Explainable AI for SSD Failure Prediction: Using LIME and SHAP for Transparency. Journal of Engineering Research and Sciences, 5(4), 1–16. https://doi.org/10.55708/js0504001